The proposed study is designed to advance understanding of risky driving, including drink-driving and drug-driving, among adolescents and young adults. The primary objective of the research is to determine the extent to which these behaviors are part of a larger lifestyle reflecting behavioral and psychosocial conventionality-unconventionality. The research will establish the location of drink-driving and drug-driving in the larger structure of risky driving practices, such as following too closely, speeding, and unsafe passing; it will determine the relationship of such risky driving practices to a variety of other problem behaviors, including problem drinking, use of marijuana and other illicit drugs, and delinquent-type behavior; and it will examine the relationship of risky driving in general, and drink- driving and drug-driving in particular, to conventionality-- unconventionality as represented by the variables of Problem- Behavior Theory, and by measures reflecting risk-taking propensity and a general orientation to hostility/aggression. The research will also examine age and gender differences in the relationship of drink-driving and drug-driving to conventionality- unconventionality. Measures of risky driving behavior, other problem behaviors, and psychosocial conventionality-unconventionality will be obtained by questionnaires mailed to 1,350 licensed Colorado drivers aged 16- 24. Participants will be selected by stratified random sampling techniques from the Colorado State Division of Motor Vehicles statewide driver database. Two separate but parallel sampling frames will be used to establish two completely independent samples that can be used to provide replications and confirmations of the analyses. In addition, official driver record data will be used to validate the self-report driving data and to serve as dependent variables to replicate analyses using self-reports. Both univariate and multivariate methods will be used in the data analyses, including correlational analysis, multiple regression analysis, structural equation modeling techniques, and confirmatory factor analysis.